There has been an exodus of sorts of mathematical expertise out of ratings agencies and into the biggest banks themselves. That may not sound like much of a change, after all it was Wall Street that hired the most Ivy League mathematicians during the last bubbles of the 1990’s and 2000’s, but the talent being sought recently is specific. Credit modeling used to be the sole purview of Moody’s, S&P and Fitch but regulatory changes, and actual blaming, are driving those kinds of activities inside banks themselves.
Since 2008, more than 300 analysts have left the major ratings companies for jobs at banks and other debt issuers, according to U.S. Securities and Exchange Commission data. Last year alone, more than 80 people made the switch, the most since the SEC began compiling such data in 2006. That’s out of a total of about 4,000 analysts employed by the ratings firms, according to SEC data.
Dodd-Frank specifically cited external ratings as a major contributor to the 2008 panic (of banks and among banks). This “mispricing” of credit led to all manner of liquidity problems that spilled out into the open, obliterating any sense of Bernanke’s “contained” sentiments. Apparently, as more new bank regulations are about to take effect next month, there is a world of difference in the mind of regulators between external credit rating agencies running models of individual credit ratings and Wall Street banks themselves running credit rating models internally.
Regulators consider the outsourcing of credit assessments to external agencies to be an “unacceptable weakening of market discipline”, said Paul Sharma, a former UK regulator now at consultants Alvarez & Marsal.
“Dodd Frank in the US already prevents banks there from relying on credit ratings when they assess risk-weighted assets, and the Basel Committee is now saying the rest of the world should take a step closer to the US approach,” he said.
What the migration of credit specialists from ratings agencies to the banks shows, however, is that it is highly likely the only manner that will change is this external/internal flip. In other words, by importing the same people running models that got it so wrong last time, banks are going to be using essentially the same models (that’s why they hired all these ratings analysts) just internally where nobody can actually see them. This is regulatory “improvement.”
The banishment of external ratings is not the only change, even specifically related to the very important task of risk-weighting. Aligning more with Basel III thinking, the Federal Reserve, the FDIC and OCC published new risk-weighting guidelines in the Federal Register meaning that the rules will take effect in late March. Among the broad changes are a re-write of the “risk bucket” approach to now include more categories and corresponding weightings. Most of the changes involve mortgages and mortgage “products”, but also attempt to gather more information and update regulator and “capital” treatment for derivatives.
NOTE: the practical impact of the risk-weighting approach to bank capital is very real and important to understanding the post-crisis environment. A full discussion of the accounting and how banks operate and make resource allocation decisions can be found here.
For example, under the current rules a non-government guaranteed mortgage loan of 1-4 family structures “prudently underwritten” and of first lien status that is both occupied (by owners or renters) and current (less than 90 days past due) would qualify for a 50% risk-weighting. Any deficiencies among those criteria would place it under the much harsher, and probably more appropriate, 100% weight. Starting March 25, those two mortgage categories will be split into several additional weighting factors, primarily related to Loan-to-Value: Category 1 loans will be 35%, 50%, 75% and 100% weightings for 60%, 80%, 90% and > 90% LTV’s, respectively. The weights for Category 2 loans will also follow an LTV breakout.
Obviously, regulators are attempting to legislate the last crisis. They have taken relevant data from 2007 and 2008 and are using that to update the regulatory model for 2015 and beyond. Staleness is a constant concept in regulations, but in this case it is much more prurient. Banks have already figured out, typically, likely pathways and loopholes around new rules usually before they hit the Federal Register. That is because they both help write the rules (not only is Wall Street a big fan of Ivy League math, it also absorbs a huge supply of Ivy League law) and have the deep resources to map out a strategy to placate them.
But in 2015, mortgages aren’t as much of an issue and certainly not where LTV’s are going to be a major factor in the next crisis; the intrusion of actual market forces and discipline starting in August 2007 took care of that, meaning these regulations are not just late they are superfluous. To that end, the new bank rules are intent on trying to address, though less specifically, some of the unknown unknowns (to cheaply borrow a war phrase), but in a manner that is almost backward – which is actually quite expected of such processes.
Schedule RC-R which governs bank call reports, which are now being required almost universally among banks since the new rules remove most size restrictions, now says:
In general, if a particular asset, derivative contract, or off-balance sheet item has features that could place it in more than one risk category, it is assigned to the category that has the lowest risk weight. For example, a holding of a U.S. municipal revenue bond that is fully guaranteed by a U.S. bank would be assigned the 20 percent risk weight appropriate to claims guaranteed by U.S. banks, rather than the 50 percent risk weight appropriate to U.S. municipal revenue bonds.
The example they use is a municipal bond, but it gets far more complex once you step into the world of derivatives and netting; particularly if that contract is split between several “uses.” All a good in-house lawyer has to do is figure out how to legally “define” a derivative contract so that it “fits” the category of only one lower risk-weight to gain that “capital” relief for the bank. And it gets even cloudier once you get into off-balance sheet credit conversion factors, which is where a great deal of contentiousness resides.
Last Friday the Wall Street Journal reported that Deutsche Bank has failed the latest “stress tests” from the Fed, not due to any quantitatively measured shortfall in “capital” adequacy, but rather in exactly this kind of legalistic and accounting “looseness.”
The expected Fed action comes amid a tussle between the U.S. regulator and foreign banks over what the Fed views as potential vulnerabilities at overseas firms with large U.S. operations. The Fed wants European banks to adhere to stricter rules and fix flaws that could potentially destabilize a firm to the point it needs support from the U.S. central bank, as happened during the 2008 financial crisis…
Deutsche Bank is in the process of fixing flaws the Fed has identified in areas ranging from its regulatory reporting, risk control, monitoring and compliance systems, these people said. The bank has also been reprimanded by the Fed for flaws in its regulatory reporting, The Wall Street Journal reported last year.
Deutsche Bank is the world’s largest purveyor of derivatives, including and especially interest rate swaps that form the basis of pretty much all credit “flow” in 2015 – a factor that I don’t see considered anywhere in the updated rules as, again, regulations are just now catching up to 2007. However, even with that patchwork understanding of 2007, there was a great deal of the wholesale system as it is now at work then, which means that the current round of regulatory “updating” is only, at best, partial.
At worst, this is all just a summary waste of time and resources. There is no way to practically regulate “risk” under the conditions of repression. Banks are under enormous pressure to take enormous risk at the same time they are getting more heavily regulated against it, so they end up doing so in the most hidden, esoteric and complex manner possible. In the end, as you might be able to discern just from the language here, all this newfound complexity in regulation serves not to make banks less “risky” but rather only to give them move cover to obscure it all under a mountain of legalisms and bank operation jargon that is impenetrable to most regulators themselves let alone the public that has the largest stake in all this (TBTF). That is the primary lesson that regulators should have learned from 2008!
Revised Part II of Schedule RC-R also includes items for reporting on- and off-balance sheet securitization exposures (items 9.a through 9.d and item 10, respectively, of Part II) separately from all other exposures subject to risk weighting. As a consequence, all securitization exposures held as on-balance sheet assets are to be excluded from the asset items 1 through 8 of Part II. Similarly, all derivatives, off-balance sheet items, and other items subject to risk-weighting that are securitization exposures are to be excluded from items 12 through 21 of Part II. In general, mortgage-backed and asset-backed securities that involve the tranching of credit risk are considered securitization exposures under the revised regulatory capital rules. For purposes of risk weighting securitization exposures, there are three different approaches from which an institution must choose (Columns Q, T, and U of Part II).
This has led to a cottage industry of “expert” consultants sprinkling seminars all over the United States to aid banks themselves (the smaller entities that don’t have the deep pockets to comply, and especially not the lobbyists to write any of the rules) to comply. So not , only do these regulations amplify the complexity, they again favor banks by size, making it far more difficult, and far less likely, for smaller banks to become bigger banks. The largest just stay at the top and gain size and scope without any practical check nor any regulatory limitation as they are well-versed in this game, having “won” the intellectual arms race with regulators and certainly legislators.
That is just one example, and it doesn’t include the proposed changes to repo positions that were actually rewritten under protest of double counting – which the dissenting comments were quite right about, meaning regulators still exhibit a shocking lack of actual understanding of these “higher order” mechanics. And that is entirely the point, as it gets lost in the minutiae of all this nit-picking by governments looking to pass off true responsibility as if more rules were the answer to every financial imbalance. The reason this “higher order” stuff all exists to begin with is both the regulations in the first place and the repression of market rates that make bank “capital” so expensive to begin with.
The eurodollar market itself owes its very existence to banks seeking regulatory relief in the 1960’s. And that may be the wholesale problem left unmoved by even these new and more rigid complexities, as they only apply to domestic operations. Though the world runs on a “dollar” standard, these US rules will not touch eurodollar operations which are far more of the wholesale manner than anything discussed here. If there is to be a tightened approach to the “dollar” it would have to be on the treatment of “dollars” as full and substitutable liabilities between domestic banks, to which these rules will apply, and their foreign “dollar” subsidiary operations, to which they won’t.
If these turn out to be too onerous, banks will, like their derivative businesses already, simply shift operations to the global “dollar” market and away from the domestic “dollar” market. At least there is an attempt, under what is being called the “Standardized Model” to harmonize bank regulations everywhere, but there will always be some kind of outlet in which to elude the endeavor. Again, the problem is not one that can be addressed by heightened rulemaking, even if it were uniform and broad, but rather in the impetus itself whereby banks seek to gain the most risk possible and hide it all – monetary repression.